No abstract

No abstract

Correctly modeling snow is critical for climate models and for hydrologic applications. Snowpack simulated by six land surface models (LSM: Noah, Variable Infiltration Capacity, snow-atmosphere-soil transfer, Land Ecosystem-Atmosphere Feedback, Noah with Multiparameterization, and Community Land Model) were evaluated against 1 year snow water equivalent (SWE) data at 112 Snow Telemetry (SNOTEL) sites in the Colorado River Headwaters region and 4 year flux tower data at two AmeriFlux sites. All models captured the main characteristics of the seasonal SWE evolution fairly well at 112 SNOTEL sites. No single model performed the best to capture the combined features of the peak SWE, the timing of peak SWE, and the length of snow season. Evaluating only simulated SWE is deceiving and does not reveal critical deficiencies in models, because the models could produce similar SWE for starkly different reasons. Sensitivity experiments revealed that the models responded differently to variations of forest coverage. The treatment of snow albedo and its cascading effects on surface energy deficit, surface temperature, stability correction, and turbulent fluxes was a major intermodel discrepancy. Six LSMs substantially overestimated (underestimated) radiative flux (heat flux), a crucial deficiency in representing winter land-atmosphere feedback in coupled weather and climate models. Results showed significant intermodel differences in snowmelt efficiency and sublimation efficiency, and models with high rate of snow accumulation and melt were able to reproduce the observed seasonal evolution of SWE. This study highlights that the parameterization of cascading effects of snow albedo and below-canopy turbulence and radiation transfer is critical not only for SWE simulation but also for correctly capturing the winter land-atmosphere interactions.

Abstract.A state-of-the-art regional model, the Weather Research and Forecasting (WRF) model (Skamarock et al., 2008) coupled with a chemistry component (Chem) (Grell et al., 2005), is coupled with the snow, ice, and aerosol radiative (SNICAR) model that includes the most sophisticated representation of snow metamorphism processes available for climate study. The coupled model is used to simulate black carbon (BC) and dust concentrations and their radiative forcing in seasonal snow over North China in January-February of 2010, with extensive field measurements used to evaluate the model performance. In general, the model simulated spatial variability of BC and dust mass concentrations in the top snow layer (hereafter BCS and DSTS, respectively) are consistent with observations. The model generally moderately underestimates BCS in the clean regions but significantly overestimates BCS in some polluted regions. Most model results fall within the uncertainty ranges of observations. The simulated BCS and DSTS are highest with > 5000 ng g −1 and up to 5 mg g −1 , respectively, over the source regions and reduce to < 50 ng g −1 and < 1 µg g −1 , respectively, in the remote regions. BCS and DSTS introduce a similar magnitude of radiative warming (∼ 10 W m −2 ) in the snowpack, which is comparable to the magnitude of surface radiative cooling due to BC and dust in the atmosphere. This study represents an effort in using a regional modeling framework to simulate BC and dust and their direct radiative forcing in snowpack.Although a variety of observational data sets have been used to attribute model biases, some uncertainties in the results remain, which highlights the need for more observations, particularly concurrent measurements of atmospheric and snow aerosols and the deposition fluxes of aerosols, in future campaigns.